4

我试图检测用户何时暂停并再次开始说话而不会停止识别。这是我设置的用于检测用户何时说话并将其输出到页面的代码:

//set up vars
var final_transcript;
var recognizing = false;
//check if using chrome and up to date
if ('webkitSpeechRecognition' in window) {
//init
  var recognition = new webkitSpeechRecognition();

  recognition.continuous = true;
  recognition.interimResults = true;

  recognition.onstart = function() {
    recognizing = true;
  };
//if there is error somewhere
  recognition.onerror = function(event) {
    console.log(event.error);
  };


  recognition.onend = function() {
    recognizing = false;
  };
//after giving the spech
  recognition.onresult = function(event) {
    var interim_transcript = '';
    for (var i = event.resultIndex; i < event.results.length; ++i) {
      if (event.results[i].isFinal) {
        // finilize and show the compleated text
        final_transcript += event.results[i][0].transcript;
      } else {
        // run the speech and output it 
        interim_transcript += event.results[i][0].transcript;

      }
    }
    final_transcript = capitalize(final_transcript);
    final_span.innerHTML = linebreak(final_transcript);
    interim_span.innerHTML = linebreak(interim_transcript);

  };
}

var two_line = /\n\n/g;
var one_line = /\n/g;
function linebreak(s) {
  return s.replace(two_line, '<p></p>').replace(one_line, '<br>');
}

function capitalize(s) {
  return s.replace(s.substr(0,1), function(m) { return m.toUpperCase(); });
}

function startDictation(event) {
  if (recognizing) {
    recognition.stop();
    return;
  }
  final_transcript = '';
  recognition.lang = 'en';
  recognition.start();
}
//startDictation();
<div id="results">
        <span id="final_span" class="final"></span>
        <span id="interim_span" class="interim"></span>

    </div>

那么有没有办法检测用户何时暂停以及他暂停的时间长度?

4

1 回答 1

3

这不是一个完整的答案,但它是开始。

https://jsfiddle.net/persianturtle/7uygdyy1/1/

我不确定该onspeechstart事件是否只触发一次,或者如果我在噪音较小的区域它是否会再次触发。

如果它多次触发,这将变得非常容易,因为我们可以从那里存储speech startspeech end计算并计算出静默时间。

如果onspeechstart它只触发一次,那么您可能会找到一种方法来获取给定文本的平均语音持续时间并从中找出静音。

但是,每次捕获单词时都会触发一个事件:onresult.

因此,基本思想是定义一个数组activity,您可以将push语音数据放入其中,然后分析静默期。

代码:

let activity = []

recognition.onresult = event => {
  console.log(event)
  activity.push('Ended:' + event.timeStamp)
}

recognition.onspeechstart = event => {
  activity.push('Started:' + event.timeStamp)
}

recognition.onend = event => {
  console.log(activity)
}
于 2017-03-11T20:51:36.747 回答